Project Summary Current electrode based approaches for recording of brain activity, as well as EEG and MEG, provide high temporal resolution, but are limited to thousands of channels. Functional MRI can interrogate hundreds of thousands of brain voxels in parallel, but is limited by hemodynamics to a temporal resolution on the order of seconds. There are no current methods that can simultaneously achieve high temporal resolution (ms scale) and high channel count (~106 channels). We propose to explore the use of microelectronic devices in conjunction with MRI, to achieve at least two orders of magnitude increase in combined spatial and temporal resolution over existing methods. In our approach large numbers of small devices on the scale of 4-500?m will be distributed across the brain, either across the surface of the brain or through the blood, and used to sense the local neural activity through electrical detection. These devices will then actively perturb their local magnetic environment in a manner that is both detectable and localizable through MR imaging. This approach combines the high temporal resolution of electrical or magnetic detection with the high spatial resolution/coverage of MRI. Two central novel concepts in this proposal are: 1) that microdevices may be used to detect local signals and output only local perturbations, rather than communicate directly with external receivers; and 2) that MRI may be an efficient means of both relaying and localizing these signals. In the long term, we believe this approach may be scalable to millions of devices on the 4?m scale circulating in the blood, providing volumetric detection of electrical activity throughout the brain. As an intermediate goal we aim to record neural activity at sub-millimeter scale and 10ms resolution from the entire cortical surface, providing a measure of electrical activity at every cortical column simultaneously.